After Hurricane Katrina ravaged the U.S. gulf coast last year, Crawford & Co., an Atlanta-based firm that provides insurance companies with outsourced claims administration services, was one of the first on the scene with eyes to assess the damage and the technological means to communicate what they saw. Crawford’s dedicated catastrophe team has adjusters ready to drop into disaster sites at a moment’s notice, along with shipping containers pre-packed with everything needed to set up claims adjusting field stations, including laptops, satellite phones and wireless LANs.
“In the middle of a disaster site, we’re probably able to be better equipped and better able to communicate than the U.S. Army,” says Glenn Gibson, CEO of Crawford’s Kitchener, Ont.-based Canadian, Caribbean and Latin American operations.
Crawford’s innovative use of IT and communications technology to speed processing of claims for disaster victims is just one example of how the insurance industry is using technology to help it deal with a growing number of increasingly costly natural and man-made disasters. From underwriting to claims processing, technology is transforming the way the industry copes – and that’s a good thing.
Insurance industry pay-outs on disaster-related claims double every five to seven years, says Paul Kovacs, a professor of Economics at the University of Western Ontario and executive director of the industry-funded Institute for Catastrophic Loss Reduction (ICLR). More people live in dangerous areas, Kovacs points out. Public infrastructure, such as storm sewers, is deteriorating. And the climate is changing.
“A relatively small change in general weather conditions produces large changes in extreme weather,” Kovacs explains. “A half a degree warmer and you get several times the number of major storms. And it looks like it will get worse from here.” Hurricane Andrew, the 1992 storm that devastated south Florida and drove more than one small insurance company into bankruptcy, was a wake-up call for the industry worldwide, Kovacs says. “Since ‘92 there have been many more and larger events, and insurance companies are paying out a lot more money, but the number of insurance companies declaring insolvency is going down.”
One thing insurance companies are learning to do – albeit in some cases slowly – is spread the risk. From a pure sales perspective, it might look like a good thing for property insurers to gain strong market share in a region or neighbourhood. But if that region is then hit by a natural disaster and all or most of your policy holders are affected, the deficit between dividends and claims could put you in a loss position.
“Companies that didn’t know from a geographical perspective where their risk was concentrated were often hit pretty hard [by major storms],” says consultant Piyush Bhatnagar, a partner in Accenture’s Toronto practice. “That in turn triggered an understanding of the need for geographic diversification of risk.”
Information technology plays a crucial role in managing that diversification. First, you need data on where you’re providing coverage. Then you need to know how that maps against risk factors – where storms or forest fires are likely to hit, for example, where municipal storm sewers or fire fighting resources are deficient. And finally you need tools to analyze and model the data.
“Most companies have some form of the data,” Bhatnagar says. “But they don’t necessarily have the analytical tools.” It’s not just a question of buying a software tool, though, he stresses. Too many companies don’t yet have the right understanding of the problem or the right business strategies or actuarial approaches in place. They need those first.
Catastrophe modeling software for the insurance industry does exist. The Dominion of Canada General Insurance Company, one of the country’s largest property and casualty insurers, uses a product from EQECAT Inc., a U.S.-based firm. Dominion doesn’t rely on just one model or tool, though. “We use three service providers and test on different models,” says Jerry Dalla Corte, the company’s vice president of business process delivery and information.
Much of the industry’s focus in this area has been around storms. Crawford and other global industry players worked with academic and government meteorologists to develop the UK-based Tropical Storm Risk service which provides Web-based tracking and year-ahead predictions of the number and strength of storms expected. It helps insurance companies model risk and allows claims departments to plan disaster responses.
“We already have the bulletin for next year,” Gibson says. “In the past it has been very accurate, surprisingly accurate.” Of course, Katrina demonstrated how easy it is to underestimate the scope of devastation from a single storm.
Most property insurance underwriters use decades-old actuarial principles based largely on historical data to calculate premiums. But the industry is beginning to realize this doesn’t make such good sense for low-probability, high-consequence events like earthquakes. Figuring out alternative methods again requires complex data modeling. “Where [catastrophe modeling] is used most prevalently is in figuring out how much exposure we have to earthquake risks,” Dominion’s Dalla Corte notes.
But like Bhatnagar, Kovacs stresses the need for business process change first. “It’s not necessarily a new machine or a new piece of software that’s needed most, but a critical change in the process of how insurance companies come up with prices,” he says.
Modernizing underwriting practices to adapt to a more disaster-prone market is well under way. Meanwhile, the claims processing end of the industry is also starting to make dramatic use of information technology. Crawford, the largest claims administration company in the world, is leading the way. Most of its effort is aimed at speeding response in disasters.
Policy holders will no longer tolerate slow service, Gibson says. They want and, many times, need money quickly to cover expenses. “Our ability to get on the ground and quickly assess losses and risk is critical to getting cash into the hands of policy holders when they need it,” he says. “Time is really of the essence.” Not just for policy holders either, it’s also important in many cases to get adjusters on the scene quickly cases so they can take steps to prevent further damage – in sewer back-ups due to flooding, for example.
Crawford’s Canadian adjusters are already using BlackBerrys so head office can reach them quickly and direct them to where they’re needed most. Some U.S. insurance companies are using GPS to track the location of adjusters so they can find the nearest one to a policy holder and dispatch them quickly. Crawford is looking at going the same route.
As in any industry, the evolution of technology proceeds along multiple paths – it’s by no means all about responding to catastrophes. Many observers say Canadian insurance companies lag their U.S. counterparts in terms of effective use of IT. Dalla Corte insists this has more to do with scale: bigger insurance companies are generally better at exploiting technology and most U.S. companies are bigger than their Canadian counterparts.
But Bhatnagar says it’s more than that. “One thing too many Canadian companies have not addressed is, ‘Are we getting value for our IT investment?’” he says. Most have not done a good job aligning business goals with IT strategies. And because many have grown recently through mergers and acquisitions, they’re saddled with disparate, often duplicate systems that don’t talk to each other and are difficult to manage. The result is they end up spending too much on “business-as-usual” IT requirements.
Simplification and consolidation of systems can free up four to five per cent of IT budgets for discretionary spending on initiatives that actually result in productivity increases. It’s an impact storm-battered insurance companies should welcome.